Optimizing Vehicle Body Cross-Sections Using a Parametric Mathematical Model

Abstract

This paper proposes a fast optimization method of body section at the conceptual design stage, based on the demand for body performance in body concept design. The study first establishes a geometrically simplified model of the truss body structure and uses the transfer matrix method to establish a fully parameterized model of the geometrically simplified body under bending conditions. Then, the stochastic gradient genetic algorithm is used to optimize the solution and determine the geometric parameters of each section. In the example of this paper, after the optimization of the established meshless model, the mass of the whole vehicle is reduced by 30%, and the stiffness of the whole vehicle is greater than that of the benchmark vehicle (5128 N/mm, 4386 N/mm), and at the same time, compared with the conceptual design method of the body of CAE technology, the modeling time is greatly reduced, and the computational efficiency of the analytical method is greatly improved compared with the finite element method.

Keywords

• conceptual design
• transfer matrix method
• mathematical modeling of body
• stochastic gradient genetic algorithm